Development Diary #8 - 3rd of December 2008

[Modestus]

I really will have to stop getting confused every time I read a Dev diary and the subsequent posts.

I understand the command structure as such......

Each Division has a general, each Corp has a general, each Army has a general, each Army Group has a general and finally each Theater has a general. Obviously your rank will be important so I imagine that an army group would need a Field marshal.

The difference from Doomsday I presume is that once you combine two divisions into a a Corp those divisions don't lose their Generals and you must appoint a new Corp commander. So as the game progresses you will need to make harder decisions on who to promote or not to promote as you may want a good divisional commander in charge of your 1st PZ div rather then making him an army commander.

Am I right about this?

 

[Bullfrog]

...you will need to make harder decisions on who to promote or not to promote as you may want a good divisional commander in charge of your 1st PZ div rather then making him an army commander.

Am I right about this?

Well it will hopefully be the case that the higher ranks (HQ leaders) will actually need to be effective. So if you start the game with a bunch of great major generals, but some poor generals, you will have to promote the lower ranks eventually...or suffer the consequences of poor leadership from theatre, armygroup and army HQs.

 

[mbb]

Exactly.... I want the option. Whether it be to have my mobile corps in NW France set to push back an invading allied landing in Brest, or my UK Singapore Defense Force evacuating to the transport fleet in the port if attacked by the Japanese, or my Vladivostock defence force left counter any Japanse attack while Im busy racing towards Berlin.....

I don't understand the resistance either. Automation would be ideal for repetitive actions like patrols, anti-partisan suppression or strategic bombing.

It would be nice just to say to a corps commander: "go get that province and tell me when you're done or you get stuck".

It might help the AI as well if objectives are broken down and tighter.

I want to be able to leave some units on standby when a specific theatre isnt the main focus of my war effort, rather than have then sit there and watch the AI land in the neighbouring province and just wave at them!

Hmm... when the British landed at Gallipoli ...

 

[mbb]

Am I right about this?

That is how I understood it.

 

[Davisx3m]

Wow!

 

[Hansag]

I don't understand the resistance either. Automation would be ideal for repetitive actions like patrols, anti-partisan suppression or strategic bombing.

It would be nice just to say to a corps commander: "go get that province and tell me when you're done or you get stuck".

I agree.
It'd be nice if generals request stuff as well, or suggest things like "1.Pz is requesting CAS in its drive towards Moscow" [Accept/Reject] etc.

Or [there were 30 divisions not 3 and your 2 Pz divisions... are in trouble] "We are facing an overwhelming enemy, permission to fall back/send reinforcements/" etc.

(and it'd be nice if we are warned WHEN not AFTER an amphibious assault is on the way)

 

[Hansag]

I really will have to stop getting confused every time I read a Dev diary and the subsequent posts.

I understand the command structure as such......

Each Division has a general, each Corp has a general, each Army has a general, each Army Group has a general and finally each Theater has a general. Obviously your rank will be important so I imagine that an army group would need a Field marshal.

The difference from Doomsday I presume is that once you combine two divisions into a a Corp those divisions don't lose their Generals and you must appoint a new Corp commander. So as the game progresses you will need to make harder decisions on who to promote or not to promote as you may want a good divisional commander in charge of your 1st PZ div rather then making him an army commander.

Am I right about this?

Yes, and from re-reading the DD, Corps HQ and above show up as units as well.

Speculation: A thought though, with a CoC system, you aught to be able to prioritize upgrades and supplies for specific theatres, then when they have calmed down, you can give this priority to another one.

 

[Razor]

Very nice!

and Yes, please inform us about supplies the next time

 

[unmerged(55467)]

Might seem obvious, but just to confirm, there are going to be NATO style icons as well as sprites for units, right?

 

[Piggy]

Might seem obvious, but just to confirm, there are going to be NATO style icons as well as sprites for units, right?

Yes.

From DD#4:

Another little aspect of the map.. If you play in "counters-mode".. the camera is locked in full 2d and will not tilt for more dramatic 3d looks.

 

[Modestus]

Well it will hopefully be the case that the higher ranks (HQ leaders) will actually need to be effective. So if you start the game with a bunch of great major generals, but some poor generals, you will have to promote the lower ranks eventually...or suffer the consequences of poor leadership from theatre, armygroup and army HQs.

Yes indeed, the idea of building the type of division you want from brigades then appointing a suitable General and the further need to manage that generals career wisely sounds very interesting.

I presume leaders will learn from experiences?

 

[WarDog]

what is it that you people hate about OPTIONAL stuff? who said you´d be forced to relinquish even a single division from your control?

Well said!


As usually Johan gives us something to dream about. Real chain of command is a huge dream come true!

I hope this will include that generic leaders will receive experience points and have the possibility to be promoted when they reach level 1 or higher.

 

[Piggy]

The fact that this question has been asked repeatedly and never answered means the developers
1) Don't know yet
2) Do know but won't tell us
or
3) No you will not be able to period!

I am leaning to the latter because if it was possible they would have answered by now. Why keep it a secret?

So unless contradicted you can safely assume what you build you are stuck with

I can only confirm that it is either 1,2 or 3.

Just plain evil....

However, Im going with #1 now... something tells me they're working on this due to feedback, perhaps.

We can hope cant we.

 

[jmschaub]

The terrain looks nice and so do the sprites... unlike others "/rant" I realize that the last step is adding the other candy (trees, etc). And this is "Alpha" guys not released software so chill. :wacko:

 

[unmerged(62754)]

Will there be some form of schematic overview of the command structure? It would be very helpful IMO to be able to view your army groups, armies, corps etc. in a fashion similar to a family tree, and even better if you could actually rearrange your forces from within this screen (in a drag-and-drop manner like a node based software).

 

[unmerged(31881)]

what is it that you people hate about OPTIONAL stuff? who said you´d be forced to relinquish even a single division from your control?

i'm quite in favour of optional stuff.

To those who are leery of it, just remember that nobody wants to make you do anything. It'd just be nice for others to have the choice available. As examples: in single-player, nobody is forced to enable Auto-promote. In multi-player, i presume it can be agreed upon. And who knows, give the options a try, you might find them worthwhile? i know i'm looking forward to them myself.

 

[Alexander Seil]

There's a bit of a contradiction in demanding micromanagement AND a decent AI simultaneously. Unless there are MASSIVE coordination penalties for microing divisions all the time, the player has an inherent advantage over the AI. Giving incentives for the players to use Corps-level AI means that the player has less of an edge, and the game becomes more challenging. Believe it or not, but, just like the AI, sometimes generals do REALLY STUPID THINGS. That's a fact of life everyone clamoring for "better AI" and "realism" on this forum always forget to mention. (I could also go on about the fact that in real life you don't get to reload, either, or play the game for 50 times until you get it perfect.)

 

[Radu]

1)I don't think a human can possibly overwhelm an AI in terms of micromanagement. Seriously,the AI sees everything it needs all the time,doesn't forget stuff,can push and click 1000 commands/second,etc.

You must have stayed away from the RTS scene for the past 5 years because the AI can actually throw a good mix of units and micromanage them better than any human with 1 brain,a pair of eyes and a keyboard and mouse.

2)Tying up the hands of players but allow the AI to do the exact same thing in the name of "balance" not only violates game design common sense (double standards) but it stopped being in fashion back in the '90s when the Civilization AI cheated like nobody's business.

3) Yes,the AI will be the single most terrible challenge the dev team will face.

 

[Alexander Seil]

1) You must have never realized you can PAUSE in HoI, right? You can't pause in Dawn of War and set everything up ten steps ahead. The human can obviously outsmart the Corps AI.

2) It's not double standards. The strategic-level AI counterpart in the other country has to rely on the same Corps AI as you do, there's no difference. Besides, HoI2 AI cheats like it's nobody's business, too. You know it has no Fog of War, right?

3) As I said, breaking it down into tasks via the command structure and orders system makes it a lot easier to conceptualize than when they worked with HoI2's clunky "fronts."

 

[Radu]

But the human cannot see what the AI does. That the AI is braindead and unable to peer into the motives of the player (encirclement,push for key provinces or good old attrition),is,like I said, the most effort-consuming correction task the devs face and the only reason why planning ahead works. Because the human knows that it faces a reactive enemy,shuffling troops about, at best and a completely passive one at worst.

A half-decent AI like the one in Operational Art III can actually make things very hard for the player.

But the key for the AI is to explicitly explain and correctly represent the important things in its universe : the game world. It should know what is important and why . This will give it an opening into the inference engine with which it could use to peer into the motives of the adversary (he's massing troops near Stalingrad - is he trying to encircle me?) . If Norm Koger's Operation Art series got the job done then it is possible.

EDIT : There's actually one thing the AI absolutely needs. Feedback from losses. He needs to learn from mistakes :

"I am UK. My convoys are getting sunk => prioritise research,builds and missions accordingly"
"I am Germany.My troops are suffering from overstretched logistics => pull back to ease logistics and attempt more encirclements"